Data Modeling. So hot right now!

#DataModeling #DataScience #DataGovernance #DataQuality #DataTrends

It can be tempting to believe that there are fashions in the tech industry. That approaches like Agile come into fashion, then go out of fashion, based on whims and whimsy. There can be an element of this:

'That other CEO has this wonderful new shiny thing - I must have it too!'

But I have found that generally businesses only do things for two reasons:

1. To try to make more money
2. Because they have no choice - they have to

And that these drivers can very often result in organisations trying to solve a problem by starting at the wrong point.

I have spent my career helping to understand and solve complex problems - and these problems have always - to a greater or lesser degree (normally greater) involved data.

Data is the input, the subject and the output of every single business process. You may think your company produces widgets, and it may, but as part of that it also ingests, processes and outputs data - and the larger the business the more data.

And if rather than producing a valuable physical thing like a widget your business provides a service - then you may not produce a tangible product at all - your product may be data!

So, data is - in a very real sense - an incredibly important and valuable asset.

To try to make more money businesses tried to extract greater value from their data, and suddenly we found that....

Everyone wanted a Data Scientist - many people wanted to be Data Scientists. But... what our valuable and skilled Data Scientists spent most of their time doing, rather than gleaning valuable insights from data, was trying to make sense of poor quality and poorly understood data and to somehow stich it all together.

So we need to fix the data quality everyone agreed, so we need lots and lots of...

Data Owners, Data Stewards, Data Quality Dashboards, Data Metrics. Massive, very expensive, master data management systems. And regulatory pressures also came in here too. Many industries must now prove they are governing and protecting their data assets - there is no longer a choice!

But, to govern, as every single colonial power ever, understood very well, you need to understand. From Rome, to the Norman Domesday Book, to the mapping and census activities of the British Empire, and every other colonial power in-between, before and after. You need to know what's there! You need a map!

I spent two wonderful summer holidays sailing in the Ionian Sea using charts that had originaly been produced by the Royal Navy in the late 18th and early 19th centuries. They didn't do this out of altruism!

And you would not try and govern an organisation without at least a basic understanding of where people were based, who they worked for, what they did, how their work was connected.

So here we are. Data is an incredibly valuable asset - in fact for many companies it is their only output.

We want to exploit this incredibly valuable asset, but it's quality is rubbish and we don't understand it.

So we need to govern our data to improve the quality (and to keep it safe! It's very valuable!). But we don't understand it! Where does it come from? Where is it now? Where does it go? How is it structured? What is it even called, and what on earth do all these different terms mean?

So, we reach today and yes at this moment, Data Modeling, so hot right now!

But doing any of these activities; exploitation, governance, modelling, in isolation misses the trick. They are all closely linked and overlap. Each depends on and feeds the others. As part of a data modeling project I uncover reams of valuable meta data and business terminology that can be part of the fuel that drives data governance. The aims of our data governors and data scientists can inform and guide the development of my data model - because one size does not fit all when it comes to data modeling.

So rather than splitting these disciplines up and having them work in isolation - they need to be combined and to work closely together, and I haven't even mentioned how our Data Architecture community also fits in here. We must not be seperate. We must work closely. Because we are all working towards the same end. A better understanding of our data so we can make better use of it!

So I would like to work towards:

Data - so hot right now! And always will be sizzling!

Because one thing is for sure - data is never going to go out of fashion!

My name is Ed Dupree and I help understand and solve complex data related problems. All views in this article are mine and mine alone and do not reflect those of my employer. I'm very excited at the moment about ontologies and taxonomies and their potentialy huge untapped potential in business contexts - so if you are an ontology geek please connect as I'd love to hear your thoughts and views - and I'm going to write another article on this subject soon!

Please add your thoughts and musings as comments to this article - I love to learn what other data people think and to benefit from their experience and knowledge.